I remember when I first learned about keyword research. The in thing then was formulas that used some combination of the number of monthly searches, the number of Google results, standardizing coefficients and the placement of the moon in relation to Mars.
As I gained more experience in actually doing keyword research, I realized that if your goal is to evaluate how competitive any given keyword is, the number of Google results for that given keyword has about as much relevance as the placement of the moon. I don't particularly care if there are 562,789 Google results. I'm not competing against the results on page 587. I'm competing against the 10 results on page 1. What's my chance of competing against them? Sometimes I might need to take a look at page 2 to get a better idea of how competitive they truly are (if page 2 results are at least as strong as I am or could be, but they're still not besting page 1 results - bad sign) - but my focus in any given keyword decision is the page 1 SERP.
So... how can you quickly and effectively evaluate the page one results and see if you have a chance of competing? The following method is not where you should end if you're planning the keyword strategy for the home page of a Fortune 500 site, but for a more minor site, or a blog post, or a site where SEO won't be the focal point of your strategy, it's worth the 10 minutes it takes.
What keyword are you going to check? You may know already what you want your keyword to be; you may want a few more ideas. Sign into your Google Adwords account (if you don't have one, create one - it won't cost you anything) and go to the keyword research tool (Tools and Analysis menu -> Keyword Tool).
Check through the list (you may want to export it to Excel and go through it if you have more than a ten-minute stake in your keywords) and choose keyphrases that look relevant to your audience and have a decent amount of monthly searches.
Now it's time to check the competition. Do a Google search for that keyphrase. Then use a tool like the Google Global addon to show the results in your target area and depersonalize them.
You should have ten results (give or take, depending on universal search add-ins) in front of you. Your goal is to survey them and decide whether the page you have in mind for this keyword can compete with them.
Look at the following factors:
1) are the title tags optimized (i.e. do they have the keyword in them, and is it in the beginning of the tag)?
2) Is the keyword in the snippet? (if not, or not as a phrase, that often says something about the on-page optimization)
3) are there any exact-match domains? (even now, there appears to be favoritism shown to exact match domains, although it may depend on the industry and may change at any moment)
4) link data - how many incoming links does the page have, and from how many root domains? (the SEOMoz bar is a good source for that; works much better if you have a PRO subscription, as then you can get all the link data, but even the "how many links" and the PA and DA metrics the free version gives you can be indicative)
5) are the pages homepages or internal pages? (homepages tend to be stronger and harder to compete against, especially if you're an internal page)
Those are the primary questions I ask myself whenever doing any kind of quick keyword research like this. For each of the questions, evaluate: where does my page stand in relation to the other pages in regards to this factor? If it's not equal or greater, could I get it there without much work? If it would require lots of work, do I or my client have the resources to do it?
10 minutes of keyword research can give you the data for a pretty educated decision. So what are you waiting for? Take the next ten minutes and go research the keywords for your next blog post!
Note: no, I did not use this process for the keywords for this blog post. I used a formula based on the number of Google results, the caloric values of holiday foods and the gravitational force of Io.